To advance our understanding of effective instructional methods for adults, we propose a five-year research program on adult literacy decoding instruction, drawing on both successful methods developed for K-12 reading instruction and successful adult literacy programs that emphasize component skills. The research proposed here will contribute to our basic knowledge on the types and amounts of decoding ability required to reach a self-sustaining level of reading fluency, and effective methods for teaching this skill to different types of adult learners. Within decoding instruction, we propose to focus on three specific issues, viz.: 1. How do rule based (deductive/abstract) and exemplar based (inductive/analogical) instructional methods affect adult learners' decoding skills? 2. How do successive and concurrent presentations of alternative (variant) pronunciations of the same graphemic pattern affect adult learners' decoding skills? 3. How does speeded practice affect adult learners' decoding skills? The research strategies adopted here include design studies (also called design experiments) followed by field studies. The design studies (Years 1-2) will be carried out in four adult literacy classrooms and will be used to test instructional approaches designed around the decoding issues. From these design experiments an enriched decoding approach will be developed and compared in a field study (Years 3-4) to existing decoding approaches and to a comparison group of non-structured instructional programs. The results will be analyzed and documented in Year 5.